Promoting Amine-Activated Electrochemical CO2 Conversion with

Subscriber access provided by Caltech Library

C: Energy Conversion and Storage; Energy and Charge Transport 2

Promoting Amine-Activated Electrochemical CO Conversion with Alkali Salts Aliza Khurram, Lifu Yan, Yuming Yin, Lingling Zhao, and Betar M Gallant J. Phys. Chem. C, Just Accepted Manuscript • DOI: 10.1021/acs.jpcc.9b04258 • Publication Date (Web): 28 Jun 2019 Downloaded from pubs.acs.org on July 21, 2019

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 26 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The Journal of Physical Chemistry

Promoting Amine-Activated Electrochemical CO2 Conversion with Alkali Salts Aliza Khurram1, Lifu Yan2, Yuming Yin2, Lingling Zhao2, Betar M. Gallant1*

1

Department of Mechanical Engineering, Massachusetts Institute of Technology, 77 Massachusetts Avenue, Cambridge, MA USA

2

Key Laboratory of Energy Thermal Conversion and Control of Ministry of Education, School of Energy & Environment, Southeast University, Nanjing 210096, China *Email: [email protected]

ABSTRACT Amine-based CO2 chemisorption has been a longstanding motif under development for CO2 capture applications, but large energy penalties are required to thermally cleave the N-C bond and regenerate CO2 for subsequent storage or utilization. Instead, it is attractive to be able to directly perform electrochemical reactions on the amine solutions with loaded CO2. We recently found that such a process is viable in dimethyl sulfoxide (DMSO) if an exogenous Li-based salt is present, leading to formation of CO2-derived products through electrochemical N-C bond cleavage. However, the detailed influence of the salt on the electrochemical reactions was not understood. Here, we investigate the role of individual electrolyte salt constituents across multiple cations and anions in DMSO to gain improved insight into the salt’s role in these complex electrolytes. While the anion appears to have minor effect, the cation is found to strongly modulate the thermochemistry of the amine-CO2 through electrostatic interactions: 1H NMR measurements show that post-capture, pre-reduction equilibrium proportions of the formed cation-associated carbamate vary by up to five-fold, and increase with the cation’s Lewis acidity (e.g. from K+ → Na+ → Li+). This trend is quantitatively supported by DFT calculations of the free energy of formation of these alkali-associated adducts. Upon electrochemical reduction, however, the current densities follow an opposing trend, with enhanced reaction rates obtained with the lowest Lewis-acidity cation, K+. Meanwhile, molecular dynamics simulations indicate significant increases in desolvation and pairing kinetics that occur with K+. These findings suggest that, in addition to strongly affecting the speciation of amine-CO2 adducts, the cation’s pairing with –COO- in the amine-CO2 adduct can significantly hinder or enhance the rates of electrochemical reactions at moderate overpotentials. Consequently, designing electrolytes to

ACS Paragon Plus Environment

1

The Journal of Physical Chemistry 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 2 of 26

promote fast cation-transfer appears important for obtaining higher current densities in future systems.

Introduction Carbon dioxide (CO2) capture and utilization (CCU) is an attractive technology to address large scale CO2 emissions from point sources such as power plants.1-2 However, CCU generally requires significant energy inputs, both upstream - during capture and concentration of CO2 at point of emission - and downstream, during subsequent reactions.3-4 For example, capture of CO2 from coal-fired plants, and its regeneration back to the gas phase, has been estimated to reduce the power output of the plant by up to 30%.5-6 Meanwhile, electrochemical conversion widely proposed as a method of downstream utilization of captured CO2, with an objective of synthesizing chemicals or fuels - involves high overpotentials associated with water splitting7-8 and the CO2 reduction reaction,9-10 further increasing energy requirements. Given these challenges, interest has been increasingly growing in combined capture-conversion processes that can directly utilize chemisorbed CO2 as the reactant in a subsequent electrochemical process, obviating the need to regenerate CO2 back to the gas phase.11-14 Utilizing chemisorbed CO2 directly, rather than post-separation CO2, invites design of new classes of electrochemical reactions by relaxing some of the key constraints of classical CO2 electroreduction.15 Given the high overpotentials required for the latter (e.g., > 1 V for CH4 on Cu electrocatalysts)16, emphasis has largely been placed on lowering the free energy barrier of the reduction intermediate, the CO2 anion (“CO2-”). This is typically accomplished through design of the electrode (catalyst) material, which can lower kinetic barriers by promoting stabilization of this high-energy species. However, significantly less attention has been devoted to modifying the reactant state of CO2 as a means to potentially improve conversion kinetics, for example by chemically incorporating CO2 into an adduct. Ideally, this species would be capable of subsequent electrochemical reactions that induce detachment of “CO2” (or a reduced CO2 anion intermediate), enabling its subsequent reactivity with electrons and cations,17 while the agent used to bind the CO2 can be re-used for further reactions. If successful, such a scheme could obviate the need for energetically-intensive CO2 release and separation between the capture and utilization steps. Realizing such a combined capture-conversion approach, however,


2

Page 3 of 26 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


requires a substrate with a CO2 binding strength that is neither too strong (impractically stabilizing the “CO2”) nor too weak. Amines, a motif abundantly studied in literature for promoting CO2 chemical interactions, are promising in this regard due to their moderate chemisorption of CO2 (-65 to -80 kJ/mol CO2) in both aqueous18-19 and nonaqueous media,20-21 which has been widely exploited for applications in CO2 capture.5, 22-24 In those applications, CO2 is typically regenerated by thermally-induced N-C bond cleavage at elevated temperatures (e.g. 100 – 120 °C),22 followed by compression for long-term storage or downstream utilization. Until recently, however, it has remained an open question whether amines can be utilized to promote direct electrochemical reactions. Table 1: Integrated chemical capture and electrochemical conversion of CO2 in a DMSObased electrolyte containing Li+-salt. CO2 capture in neat DMSO: 2RNH2 + 2CO2 → 2RNHCOOH Salt-induced re-speciation: 2RNHCOOH + Li+ → RNHCOO-Li+ + RNH3+ + CO2 Net capture step: 2RNH2 + CO2 + Li+ → RNHCOO-Li+ + RNH3+

(1) (2) (3)

Electrochemical conversion step: RNHCOO-Li+ + RNH3+ + 2CO2 + 3Li+ + 4e- → (4) 2Li2CO3 + C + 2RNH2 Net capture-conversion reaction: 3CO2 + 4Li+ + 4e- → 2Li2CO3 + C

(5)

Recently, we reported that such a scheme is possible. We studied chemisorption of CO2 by an alkylamine, 2-ethoxyethylamine (EEA), in dimethylsulfoxide (DMSO) with an exogenous Li+-based co-salt. In neat DMSO (no co-salt), CO2 uptake by EEA resulted in the preferential formation of carbamic acid (Reaction 1 in Table 1). Subsequent inclusion of the salt, however, induced conversion of all carbamic acid to lithium carbamate (Reaction 2). We found that the lithium carbamate, and not the carbamic acid, was intrinsically electroactive, exhibiting reduction activity at high potentials of ~2.8 V vs. Li/Li+. Meanwhile, physisorbed CO2 exhibited no electroactivity in the same DMSO electrolyte without the amine.25 Upon electron transfer, the carbamate species were found to undergo selective N-C bond cleavage, with the resulting products (predominantly Li2CO3) consisting of only CO2-derived species. Meanwhile, the amine remained in solution, capable of further CO2 uptake. The overall electrochemical reaction


3


Page 4 of 26

occurred according to Reaction 4, in which the amine-bound CO2, as RNHCOO-, played the central role as the electroactive species (Figure 1). Our results suggested that the product lean amine (2RNH2) could undergo limited turnovers ( Na+ > K+ > TBA+. The final equilibrium proportion of alkali carbamate formed 96 hours after salt addition was determined to vary by nearly a factor of five depending on the alkali cation alone, yielding 49.5%, 32.1%, and 11.3% carbamate for Li+, Na+, and K+, respectively. Unexpectedly, a small amount of carbamate (7.4%) formed even with TBA+, indicating a non-negligible, yet very weak, carbamate-TBA+ interaction (the TBA+ spectrum is omitted from Figure 3 due to large background 1H resonances, but time-dependent spectra are provided in Figure S5). These results clearly indicate that Li+ associates most strongly with the initial carbamic acid, driving the conversion reaction to carbamate farthest to the right in DMSO. We could also separately confirm this trend in the NMR spectra, namely that solutions containing stronger Lewis acids corresponded to lower CO2 loadings and thus higher carbamate proportions, by directly measuring the amount of CO2 bound to EEA in the different equilibrated solutions. This was


10

Page 11 of 26

accomplished by treating CO2-rich electrolytes with 2 M H3PO4,28 which released all chemicallybound CO2 back to the gas phase where it was quantified via mass spectrometry. As shown in Figure S6, the CO2 loading ratios for Li+ and K+, twenty hours after the addition of the salt, were 0.58 and 0.77 CO2 per EEA, respectively (theoretical loading ratios are 0.5 for purely carbamate and 1 for purely carbamic acid), in good agreement with 1H NMR spectra results. Note that the presence of trace water, potentially due to variations in different water contents of salts, could be ruled out as the origin of varying equilibrium proportions of carbamate, as water content in each of the as-prepared solutions was found to be negligible ( Na+ > K+, as smaller Gibbs free energy changes would favor lower populations of the product carbamate. In addition, the computed enthalpies of reaction, Hr, which range from -133.3 kJ/mol (Li+) to -73.2 kJ/mol (K+), agree well with experimental values of alkyl amines measured elsewhere in aprotic solvents without added co-salts.20 Overall, these results highlight the central role of the alkali cations in driving key shifts in equilibrium speciation of the EEA-CO2 adducts which are necessary for the subsequent electrochemical activity to be observed.


12

Page 13 of 26

(a)

RNHCOOH

RNHCOOH

O-C-O Angle

(c)

RNHCOO-

A+ DGr , DHr

RNH3+

(b)

O

C

N

CO2 H

A+

DGr (kJ/mol)

DHr (kJ/mol)

No Salt

2.5

36.0

Li+

-143.1

-133.3

Na+

-111.8

-101.3

K+

-85.7

-73.2

N-C Length (Å)

RNHCOOA

A+

A+

C-N Length

O-C-O Angle (°)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


1.40 (i) 1.38 1.36 H+

Li+

Na+

K+

H+

Li+

Na+

K+

126 (ii) 123 120

Figure 5: (a) Optimized structures of RNHCOOH and A+ (reactants) and RNHCOO-A+, RNH3+, and CO2 (products) obtained via DFT post-geometry minimization (A+ = Li+, Na+, or K+). (b) Free energy (Gr) and enthalpy (Hr) differences incurred upon 2RNHCOOH + A+ → RNHCOO-A+ + RNH3+ + CO2. All enthalpy and free energy calculations were performed at T = 298 K and PCO2 = 1 atm. (c) N-C bond lengths and O-C-O bond angles in carbamic acid (RNHCOOH) and alkali metal carbamates. The resulting N-C bond lengths and O-C-O bond angles were also found to vary significantly across the different alkali carbamates. As shown in Figure 5c, as electrostatic interactions decreased from Li+ to K+, the N-C bond lengthened monotonically from 1.38 to 1.40 Å, indicating N-C bond weakening with larger cation size. Carbamic acid (-COOH) possessed the shortest N-C bond length (1.50% shorter than in Li carbamate), and therefore the strongest N-C bond. This suggests that in comparison to alkali carbamates, neutral carbamic acid (more stable) is more difficult to electrochemically reduce via selective cleavage of the N-C bond than the alkali carbamate in solutions where both populations exist, favoring selective reduction of the latter. For alkali carbamates, the cation-dependent trend in N-C bond shortening is attributable to higher inductive effects around the terminal –COO for harder Lewis acids (e.g. Li+), resulting in lower electron density around the carboxylate carbon and bond strengthening. Correspondingly, the O-C-O bond angle in the adduct varied proportionally, with smaller O-C-O bond angles (more “bent”) for stronger Lewis acids (122.3° vs 125.1° for Li+ or K+, respectively) and shorter


13


Page 14 of 26

N-C bond lengths. The trend of N-C bond strengthening with harder Lewis acids could also be experimentally validated by ATR-IR experiments performed with 1.5 M CO2-loaded EEA and 2 M LiClO4 (concentrations were proportionally increased to accentuate the relatively weak sensitivity of N-C stretches in IR measurements). As shown in Figure S7, consistent with the DFT calculations, the N-COO- stretching vibration shifted from a higher wavenumber for Li carbamate (~1321 cm-1), to lower wavenumbers for Na (~1311 cm-1) and K (~ 1309 cm-1) carbamate.

Moreover, the characteristic carbamate peaks at 1576 and 1494 cm-1, and the

stretching vibration of N-COO- at 1321 cm-1,39-41 are strongest in overall intensity for Li+ and decrease monotonically thereafter for Na+ and K+, consistent with maximum carbamate formation with Li+. DFT calculations also enabled interrogation of the LUMO structure in the alkaliassociated carbamates, which indicate a significant degree of localization over the carbamate/cation moiety with little contribution from the alkyl ethoxy backbone (Figure S8). This suggests that under subsequent electrochemical conditions, the first elementary electron transfer step results in added electron density centered generally around the N-C bond, yet delocalized somewhat over the -N-COO-A+ terminal end. Meanwhile, the backbone, i.e. CH3CH2OCH2CH2-, is expected from these calculations to be relatively unaffected, and therefore stable, upon electron transfer. We further investigated the change in electronic structure of lithium carbamate upon the addition of a single electron to predict changes that may occur under electrochemical conditions. While the bond length changes are generally an order of magnitude larger for the bonds in the NCOO- termination than they are for those in the alkyl ethoxy tail of the amine (Figure S9), the overall changes were relatively minor (< 2%) for the N-C and C-O bonds, respectively. This suggests that although these bonds are most significantly weakened upon reduction, the reduced form of the adduct may remain relatively stable for subsequent ion transfer, electron transfer, and/or reaction with additional CO242 before the N-C bond cleavage event occurs. Effect of electrolyte salt on the electrochemical activity of CO2-loaded EEA Having characterized the electrochemical reactant state in detail, we next studied how the salt constituents influence the subsequent electrochemical reactions, in which the electrolytes are perturbed from equilibrium. To investigate the influence of the co-salt anion with Li+ as the cation, two different electrochemical configurations were compared: (1) Discharge in two-


14

Page 15 of 26

electrode cells with a Li metal anode and Vulcan carbon-based cathode, for suitable comparison with previous results;25 and (2) A three-electrode electrolysis-type cell containing a Super P carbon - coated glassy carbon working electrode, a fritted Ag/AgNO3 reference electrode and a fritted Pt counter electrode (see SI for additional details). In both setups, following electrolyte preparation, CO2 introduction, and resting (20 hours), the solutions were re-saturated with CO2 prior to each measurement to ensure a CO2 headspace. The three-electrode setup was particularly useful in later measurements to enable comparison across different cations, where stable metal anode analogues to Li are not available. E (V vs. Li/Li+) 2.22

0.000

-0.004

(b)

CO2 Only ClO4-

-0.008

PF6-

TFSI

j (mA/cm2)

(a) j (mA/cm2)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


-

Li+ 1 mV/s

-0.012 2.2

2.4

2.6

2.8

3.0

3.2

2.42

2.62

2.82

3.02

CO2 Only

0.000

TBA+ -0.004 Na+

Li+

-0.008

-0.012

K -1.40

ClO41 mV/s

+

-1.20

-1.00

-0.80

-0.60

+

+

E (V vs. Ag/Ag )

E (V vs. Li/Li )

Figure 6: Comparison of capacitance- and iR-corrected linear sweep voltammograms at 1 mV/s in 3-electrode measurements with 0.1 M EEA-CO2 and (a) 0.3 M LiY/DMSO (Y- = ClO4-, PF6-, or TFSI-) or (b) 0.3 M AClO4 (A+ = Li+, Na+, K+ or TBA+). The top axis in Figure 6b displays the potential vs. Li/Li+ for the scan corresponding to LiClO4 to facilitate comparison with prior galvanostatic results in Li cells. The working electrode in all cases was a super P-coated glassy carbon electrode, and current densities were normalized to the glassy carbon geometric surface area (0.196 cm2). Electrochemical measurements on the equilibrated solutions revealed almost identical behavior for all Li-based salts, regardless of the anion. Under galvanostatic discharge conditions in two-electrode cells, nearly identical voltage plateaus (~2.85 V vs. Li/Li+) and capacities (> 3400 mAh/gc) were observed (Figure S10) at a fixed rate of 20 mA/gc. The nearly-identical redox potentials confirm that the anion chemistry factors little in the overall free energy change during conversion, and therefore that the anion’s participation is weak or negligible in the overall electrochemical reaction. The comparable capacities reflect that the cell performance was limited by the formation of similar quantities of passivating Li2CO3 during discharge.25 Similar findings


15


Page 16 of 26

were obtained with linear sweep voltammetry measurements conducted at 1 mV/s in Figure 6a. Whereas only capacitive backgrounds were observed for either physically-dissolved CO2 (no EEA) or lean EEA (no CO2) in the neat electrolytes (Figure S11), solutions containing EEACO2 exhibited an evident redox peak upon the negative scan, with similar reduction current magnitudes obtained at comparable onset reduction potentials across the different anions (2.68, 2.69 and 2.70 V vs. Li/Li+ for ClO4-, TFSI- and PF6-, respectively). Linear sweep voltammetry (LSV) measurements were next conducted with varying cations under the same conditions. As shown in Figure 6b, nearly-identical onset potentials (approximately -0.94 V vs. Ag/Ag+) were observed with Li+, Na+ and K+ (all CV background scans are included in Figure S12). This suggests that the measured electrochemical potentials at reduction onset, in addition to being largely insensitive to anion as expected, also do not strongly reflect the different strengths of ionic interaction in the alkali carbamates. Note that the expected Nernstian shift in the thermodynamic potential of Reaction (1) due to varying concentrations of initial carbamate is less than 10 mV for the relevant range of carbamate proportions here (approximately 50 mM for Li+, 32 mM for Na+, and 11 mM for K+), and therefore differences in reactant population would not be expected to strongly affect the measurable redox potential. Although there was negligible variation in onset potential, substantially higher reduction rates were obtained with K+. A larger reduction peak current density (-11.2 μ A/cm2) was obtained with K+ compared to Li+ and Na+ (approximately -6.2 μA/cm2 for both). The exchange current density was also greatest for K+, as qualitatively determined from Tafel responses shown in Figure S13 as a function of cation. Although it was not possible to directly characterize the reduction products formed with Na+ and K+ (see Supporting Information), the nearly identical onset potentials suggest that the overall reaction scheme is not altered by the alkali cation; i.e., Reaction 4 is generalizable to Na+ and K+. Interestingly, the EEA-CO2 adducts were also slightly electrochemically active with TBA+; however, a noticeably lower onset reduction potential (by approximately 80 mV) and peak current density (-1.78 μA/cm2) were observed in comparison to the alkali cations. The electroactivity observed with TBA+ is consistent with the small fraction of TBA+ carbamate formed (Figure 4), although, it is unclear at present what reduction products are formed with TBA+ given the likely inability to form carbonate. Taken together, the combined NMR, ATR-IR, DFT, and electrochemical results indicate: (1) A clear correlation between harder Lewis acid cations and higher quantities of carbamate formed, which reflects the


16

Page 17 of 26 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


modulated chemical thermodynamics of the reactant state prior to electrochemical reactions; (2) A prevailing role of softer Lewis acid cations, namely K+, in amplifying the rate of the reduction reaction. Understanding how the cation’s selection may factor into higher reduction currents requires non-thermodynamic treatment of the cation’s transport properties to investigate more carefully. Therefore, we next employed molecular dynamics (MD) simulations to investigate the different solvation structures and transport behaviors of alkali cations as they approach, and pair with, an available carbamate in solution, simulating the likely ion-transfer process occurring under electrochemical conditions. Molecular dynamics simulations of the ion-pairing events under reaction conditions To elucidate the role of the cation under electrochemical conditions, we used MD simulations to investigate the rate and energetics of dynamic ion-pairing between the alkali cation and carbamate anion, which reflects the relative facility of ion transfer from bulk solution in DMSO to the reaction sphere. Note that ion-pairing comprises two coupled processes: loss of solvent from the cation solvation shell and around the –COO- in the amine-; and concurrent electrostatic pairing of the de-shielded alkali cation and carbamate. Our previous characterization of the reactant state indicates that one alkali cation is already pre-associated with each carbamate molecule at the onset of reduction. However, Reaction 3 indicates that three additional alkali cations are consumed in the reduction reaction. The DFT calculations also indicated that electron transfer alone did not lead to detachment of the CO2-derived species, supporting the fact that additional ion transfer steps occur to the negatively-charged carbamate. Therefore, we hypothesized that the availability of alkali cations for reaction with RNHCOO- would be an important factor in determining the electrochemical reaction rates. MD was thus used to model the following reaction in which RNHCOO- represents a model amine-derived anion: RNHCOO−(solv) + A+(solv) → RNHCOOA(solv)

(6)

where (solv) denotes a solvated species in DMSO. Note that RNHCOO- is simply used here to represent a general anionic form of the amine-CO2 adduct formed during electrochemical reactions, which may, in practice, correspond to the anion, dianion, and/or cation-paired variants of these species. The typical computational domain had dimensions of 7 × 7 × 7 nm3 and


17


Page 18 of 26

contained 2885 DMSO molecules, yielding a simulated density of 1.09 g/cm3, compared to an experimentally established value of 1.1 g/cm3 at STP. Twenty alkali cations and RNHCOO– ions each were added to the computational domain, mimicking the amine concentration of 0.1 M used in experiments. Co-salt anions and ammonium cations were not included to simplify the simulations. The equations of motion were integrated using the Verlet (Leap-Frog) algorithm with a time step of 1 fs and simulation time of 80 ns. Initial unpaired and final paired configurations for Li+ are shown in Figure 7a-b, respectively, with Na+ and K+ results in Figure S14. For method validation, the coordination numbers (CNs) of alkali cations in bulk DMSO (no amine) were first computed, based on the number of ODMSO in the cation’s first solvation shell, as 4.1, 4.8, and 5.2 for Li+, Na+, and K+, respectively (Figure 7c), consistent with literature reports (Li+: 4, Na+: 5, and K+: 6).43-45 The MD simulations indicated, as expected, that the paired configurations corresponded to partial desolvation of all cations to permit electrostatic association with carbamate. However, both the extent, and the rate, of desolvation varied significantly depending on the cation. Specifically, the pairing kinetics of K+ with RNHCOO- are significantly more rapid than with either Li+ or Na+. Out of a total of 20 possible pairs, in 80 ns, all 20 ion pairs formed for K+. Meanwhile, over the same simulation time, only eight pairs formed for Li+, and three for Na+ (Figure 7d). These rates corresponded to approximately 3, 8, and 13 pairs formed/ns, respectively (Table S2). This more rapid pairing is attributed to the weaker solvation strength of K+ in DMSO, resulting in the larger and faster change in the CN for K+ (CN change of 0.7, 0.3, and 2.2 for Li+, Na+, K+ respectively after 80 ns, Figures 7c and S15). To quantify the desolvation kinetics, potential of mean force (PMF) calculations were carried out as a function of − center-of-mass separation between A+ and -COORNHCOO in DMSO. As shown in Figure S16, the

energy barrier, Ea, to reach the transition state of the ion pairing process (Reaction 3) is the smallest for K+ (Ea=10.1 kJ/mol), followed by Li+ (Ea=13.8 kJ/mol) and Na+ (Ea=15.0 kJ/mol). Together, these results explain why higher reduction currents are obtained with K+, and suggest, interestingly, that ion-transfer from bulk solution to the amine is rate-determining at moderate overpotentials. Note that even though K+ pairs with RNHCOO- most rapidly when the carbamic acid is already deprotonated (as may occur during reaction conditions), it does not imply rapid initial formation of potassium carbamate, because as discussed earlier, the driving force to deprotonate the acid and form potassium carbamate is the lowest among the three alkali cations


18

Page 19 of 26

investigated. Thus, the beneficial energetics of desolvation and ion pairing would have the greatest effect when the system is perturbed from equilibrium and is in ionic form, i.e., actively undergoing electrochemical reactions where facile supply of cations may be essential to sustain

(a)

(c) Li+

DMSO

RNHCOO-

Unpaired

(b)

(d) DMSO

RNHCOOLi

Paired

A+-ODMSO Coordination Number

reactions, highly consistent with the experimental observations.

No. of Ion Pairs (out of 20)

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


5.5 5.0 4.5 4.0

Unpaired

3.5 3.0 2.5

Paired Li

+

K+

Na+

+ 20 RNHCOO ---A

K+

15 10

Li+

5 Na+

0 0

20

40

60

80

time (ns)

Figure 7: First solvation shell structures for Li+ in (a) an unpaired configuration, with Li+ fully solvated by DMSO, and in (b) a paired configuration, with Li+ electrostatically coordinated to RNHCOO- following a simulation time of 80 ns. (c) Coordination numbers of the alkali cation, A+, with DMSO (A+-ODMSO), computed at equilibrium with all ions initialized in either the unpaired or paired configurations. (d) Number of ion pairs (RNHCOO-A+) formed as a function of computational time, out of a total of 20 possible pairs. Taken together, these results provide, first, new and unambiguous evidence that amineCO2 adducts exhibit activity in a broader range of added co-salts than known previously. This indicates that the amine-activation process reported in our first work is more generally applicable than previously understood, opening up the possibility for further development. Second, it reveals mechanistic ways in which the salt does, and does not, play a role. The co-salt cation, and


19


Page 20 of 26

its extended solvation structure, is associated with the adduct prior to electrochemical reactions, and must therefore form a part of the definition of the electroactive reactant state. However, both the reactant state concentration and the identity of the associated alkali cation factor little in the onset reduction potentials, at least for the amine and solvent (EEA/DMSO) examined here. In contrast, the activity at lower potentials is strongly governed by the cation. Our results indicate that this can be explained by the relative availability of bulk ions to react with either carbamate or associated anionic forms, which is favored for K+, having a weaker solvation strength and lower desolvation barrier. It is interesting to note that Na+, which lies between Li+ and K+ in terms of Lewis acidity, exhibits a linear sweep response and peak current densities that are very similar to that of Li+, rather than lying in between Li+ and K+ (Figure 6b). Although elucidating this trend requires further detailed understanding of the electrochemical mechanisms, planned in future work, it is fully consistent with the MD calculations, which indicate that Na+ has the highest desolvation barrier and therefore cannot sustain high reaction rates due to impeded ion transfer. Overall, our results indicate that coupled ion-transfer is a governing step in the electrochemical reactions of amine-CO2 adducts in DMSO. Therefore, use of more weaklysolvating solvents in future work may enable higher electrochemical rates to be sustained. In addition, it will be important to identify new amine and solvent combinations with improved bulk transport properties in the CO2-loaded state. This could support both faster initial rates of CO2 capture and alkali carbamate (electroactive species) formation, which may also improve electrochemical reaction rates. Amines that undergo weaker intermolecular hydrogen bonding and viscosity changes upon CO2 loading are currently being developed for CO2 capture,46-48 and their suitability for electrochemical reactions should be explored to further support improved capture and conversion rates. If the above limitations can be addressed, the DFT results indicate that further tailoring of amine chemistry to manipulate and optimize N-C bond strength and the overall electronic structure in the EEA-CO2 adduct may enable finer control over the redox potential and, presumably, the intrinsic electrochemical kinetics.

Conclusions


20

Page 21 of 26 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


In summary, these results elucidate the critical role of alkali metal-based co-salts in inducing key shifts in amine-CO2 adduct speciation, which can be directly linked to subsequent electrochemical activity. Whereas carbamic acids (-COOH) cannot be electrochemically reduced in the absence of free ions, adding electrolyte co-salts yields cation-associated carbamates of EEA that exhibit significantly enhanced electrochemical activity compared to CO2 only, at potentials in excess of 2.7 V vs. Li/Li+ or -0.94 vs. Ag/Ag+ (nonaqueous). The carbamic acidcarbamate equilibria can be modified by altering the electrolyte co-salt cation (Li+, Na+, K+, and TBA+ investigated herein), with a smaller cation size resulting in increased carbamate formation due to increasing strength of cation – carbamate ionic interactions. Upon electrochemical reduction, however, K+-paired carbamates show the greatest reduction currents, in spite of the lowest overall population of active carbamates formed. MD simulations reveal that this effect is likely non-thermodynamic in nature, resulting from improved K+ transfer from bulk solution to the reaction site during electrochemical reduction. Although the full multi-bodied, multi-step reaction mechanisms are complex and require continued examination to elucidate in full, these results indicate that electrochemical reaction rates of amine-CO2 adducts may be significantly improved by carefully tailoring cation-solvent-amine interactions.

Conflicts of interest There are no conflicts to declare.

Acknowledgements This research was supported by startup funding from the MIT Department of Mechanical Engineering. The authors acknowledge the Small Molecule Mass Spectrometry—A Harvard FAS Division of Science Core Facility - for assistance with mass spectrometry measurements. This work made use of the MRSEC Shared Experimental Facilities at MIT, supported by the National Science Foundation under award number DMR-14-19807. L.Z. acknowledges funding from the National Natural Science Foundation of China (Grant No. 51776041) and the China Scholarship Council (201706095025). This work used the Extreme Science and Engineering Discovery Environment (XSEDE)49 (supported by National Science Foundation grant number ACI-1548562) Comet at the San Diego Supercomputer Center through allocation TGCTS180040.

Supporting Information


21


Page 22 of 26

Detailed experimental and computational methods, carbamate quantification calculation from 1H NMR, time-dependent 1H NMR spectra, Karl Fischer measurements for water content determination, linear sweep voltammograms for background (no EEA and no CO2) cases, Tafel responses, electronic density distribution for carbamates, first solvation shell structures for Na+ and K+, rate of change in coordination number, variation in potential of mean force (PMF) upon ion-pairing, and residence time correlation functions.

References 1.

2. 3.

4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14.

Markewitz, P.; Kuckshinrichs, W.; Leitner, W.; Linssen, J.; Zapp, P.; Bongartz, R.; Schreiber, A.; Müller, T. E., Worldwide innovations in the development of carbon capture technologies and the utilization of CO2. Energy Environ. Sci. 2012, 5 (6), 72817305. Hunt, A. J.; Sin, E. H. K.; Marriott, R.; Clark, J. H., Generation, Capture, and Utilization of Industrial Carbon Dioxide. ChemSusChem 2010, 3 (3), 306-322. Appel, A. M.; Bercaw, J. E.; Bocarsly, A. B.; Dobbek, H.; DuBois, D. L.; Dupuis, M.; Ferry, J. G.; Fujita, E.; Hille, R.; Kenis, P. J. A. et al., Frontiers, Opportunities, and Challenges in Biochemical and Chemical Catalysis of CO2 Fixation. Chem. Rev. 2013, 113 (8), 6621-6658. Whipple, D. T.; Kenis, P. J. A., Prospects of CO2 Utilization via Direct Heterogeneous Electrochemical Reduction. J. Phys. Chem. Lett. 2010, 1 (24), 3451-3458. Boot-Handford, M. E.; Abanades, J. C.; Anthony, E. J.; Blunt, M. J.; Brandani, S.; Mac Dowell, N.; Fernández, J. R.; Ferrari, M.-C.; Gross, R.; Hallett, J. P. et al., Carbon capture and storage update. Energy Environ. Sci. 2014, 7 (1), 130-189. Aaron, D.; Tsouris, C., Separation of CO2 from Flue Gas: A Review. Sep. Sci. Technol. 2005, 40 (1-3), 321-348. Hong, W. T.; Risch, M.; Stoerzinger, K. A.; Grimaud, A.; Suntivich, J.; Shao-Horn, Y., Toward the rational design of non-precious transition metal oxides for oxygen electrocatalysis. Energy Environ. Sci. 2015, 8 (5), 1404-1427. Walter, M. G.; Warren, E. L.; McKone, J. R.; Boettcher, S. W.; Mi, Q.; Santori, E. A.; Lewis, N. S., Solar Water Splitting Cells. Chem. Rev. 2010, 110 (11), 6446-6473. Benson, E. E.; Kubiak, C. P.; Sathrum, A. J.; Smieja, J. M., Electrocatalytic and homogeneous approaches to conversion of CO2 to liquid fuels. Chem. Soc. Rev. 2009, 38 (1), 89-99. Gattrell, M.; Gupta, N.; Co, A., A review of the aqueous electrochemical reduction of CO2 to hydrocarbons at copper. J. Electroanal. Chem. 2006, 594 (1), 1-19. Machado, A. S. R.; da Ponte, M. N., CO2 capture and electrochemical conversion. Curr. Opin. Green Sustainable Chem. 2018, 11, 86-90. Yang, Z. Z.; Zhao, Y. N.; He, L. N., CO2 chemistry: task-specific ionic liquids for CO2 capture/activation and subsequent conversion. Rsc Adv 2011, 1 (4), 545-567. Arti, M.; Youn, M. H.; Park, K. T.; Kim, H. J.; Kim, Y. E.; Jeong, S. K., Single Process for CO2 Capture and Mineralization in Various Alkanolamines Using Calcium Chloride. Energy Fuels 2017, 31 (1), 763-769. Luca, O. R.; Fenwick, A. Q., Organic reactions for the electrochemical and photochemical production of chemical fuels from CO2 - The reduction chemistry of carboxylic acids and derivatives as bent CO2 surrogates. J. Photochem. Photobiol. B 2015, 152, 26-42.


22

Page 23 of 26 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


15. 16. 17. 18. 19. 20. 21. 22. 23. 24.

25. 26. 27. 28. 29. 30.

Luca, O. R.; McCrory, C. C. L.; Dalleska, N. F.; Koval, C. A., The Selective Electrochemical Conversion of Preactivated CO2 to Methane. J. Electrochem. Soc. 2015, 162 (7), H473-H476. Kuhl, K. P.; Cave, E. R.; Abram, D. N.; Jaramillo, T. F., New insights into the electrochemical reduction of carbon dioxide on metallic copper surfaces. Energy Environ. Sci. 2012, 5 (5), 7050-7059. Chen, L.; Li, F.; Zhang, Y.; Bentley Cameron, L.; Horne, M.; Bond Alan, M.; Zhang, J., Electrochemical Reduction of Carbon Dioxide in a Monoethanolamine Capture Medium. ChemSusChem 2017, 10 (20), 4109-4118. McCann, N.; Maeder, M.; Attalla, M., Simulation of Enthalpy and Capacity of CO2 Absorption by Aqueous Amine Systems. Ind. Eng. Chem. Res. 2008, 47 (6), 2002-2009. Kim, I.; Svendsen, H. F., Heat of Absorption of Carbon Dioxide (CO2) in Monoethanolamine (MEA) and 2-(Aminoethyl)ethanolamine (AEEA) Solutions. Ind. Eng. Chem. Res. 2007, 46 (17), 5803-5809. Svensson, H.; Zejnullahu Velasco, V.; Hulteberg, C.; Karlsson, H. T., Heat of absorption of carbon dioxide in mixtures of 2-amino-2-methyl-1-propanol and organic solvents. Int. J. Greenhouse Gas Control 2014, 30, 1-8. Gurkan, B. E.; de la Fuente, J. C.; Mindrup, E. M.; Ficke, L. E.; Goodrich, B. F.; Price, E. A.; Schneider, W. F.; Brennecke, J. F., Equimolar CO2 Absorption by AnionFunctionalized Ionic Liquids. J. Am. Chem. Soc. 2010, 132 (7), 2116-2117. Rochelle, G. T., Amine scrubbing for CO2 capture. Science 2009, 325 (5948), 1652-4. Heldebrant, D. J.; Koech, P. K.; Glezakou, V. A.; Rousseau, R.; Malhotra, D.; Cantu, D. C., Water-Lean Solvents for Post-Combustion CO2 Capture: Fundamentals, Uncertainties, Opportunities, and Outlook. Chem. Rev. 2017, 117 (14), 9594-9624. Seo, S.; Quiroz-Guzman, M.; DeSilva, M. A.; Lee, T. B.; Huang, Y.; Goodrich, B. F.; Schneider, W. F.; Brennecke, J. F., Chemically Tunable Ionic Liquids with Aprotic Heterocyclic Anion (AHA) for CO2 Capture. J. Phys. Chem. B 2014, 118 (21), 57405751. Khurram, A.; He, M.; Gallant, B. M., Tailoring the Discharge Reaction in Li-CO2 Batteries through Incorporation of CO2 Capture Chemistry. Joule 2018, 2 (12), 26492666. Amatore, C.; Saveant, J. M., Mechanism and kinetic characteristics of the electrochemical reduction of carbon dioxide in media of low proton availability. J. Am. Chem. Soc. 1981, 103 (17), 5021-5023. Yang, S. X.; Qiao, Y.; He, P.; Liu, Y. J.; Cheng, Z.; Zhu, J. J.; Zhou, H. S., A reversible lithium-CO2 battery with Ru nanoparticles as a cathode catalyst. Energy Environ. Sci. 2017, 10 (4), 972-978. Bishnoi, S.; Rochelle, G. T., Absorption of carbon dioxide into aqueous piperazine: reaction kinetics, mass transfer and solubility. Chem. Eng. Sci. 2000, 55 (22), 5531-5543. Farrokhpour, H.; Manassir, M., A Simple Method for Estimating the Absolute Solvation Free Energy of Monovalent Ions in Different Solvents. J. Phys. Chem. A 2015, 119 (1), 160-171. Marenich, A. V.; Cramer, C. J.; Truhlar, D. G., Universal Solvation Model Based on Solute Electron Density and on a Continuum Model of the Solvent Defined by the Bulk Dielectric Constant and Atomic Surface Tensions. J. Phys. Chem. B 2009, 113 (18), 6378-6396.


23


31. 32. 33. 34.

35.

36. 37. 38. 39. 40. 41. 42. 43. 44. 45. 46.

Page 24 of 26

Van Der Spoel, D.; Lindahl, E.; Hess, B.; Groenhof, G.; Mark Alan, E.; Berendsen Herman, J. C., GROMACS: Fast, flexible, and free. J. Comput. Chem. 2005, 26 (16), 1701-1718. Strader, M. L.; Feller, S. E., A Flexible All-Atom Model of Dimethyl Sulfoxide for Molecular Dynamics Simulations. J. Phys. Chem. A 2002, 106 (6), 1074-1080. Jorgensen, W. L.; Maxwell, D. S.; Tirado-Rives, J., Development and Testing of the OPLS All-Atom Force Field on Conformational Energetics and Properties of Organic Liquids. J. Am. Chem. Soc. 1996, 118 (45), 11225-11236. Kortunov, P. V.; Siskin, M.; Baugh, L. S.; Calabro, D. C., In Situ Nuclear Magnetic Resonance Mechanistic Studies of Carbon Dioxide Reactions with Liquid Amines in Non-aqueous Systems: Evidence for the Formation of Carbamic Acids and Zwitterionic Species. Energy & Fuels 2015, 29 (9), 5940-5966. Kortunov, P. V.; Siskin, M.; Baugh, L. S.; Calabro, D. C., In Situ Nuclear Magnetic Resonance Mechanistic Studies of Carbon Dioxide Reactions with Liquid Amines in Aqueous Systems: New Insights on Carbon Capture Reaction Pathways. Energy & Fuels 2015, 29 (9), 5919-5939. Andreoni, W.; Pietrucci, F., CO2 capture in amine solutions: modelling and simulations with non-empirical methods. J. Phys.: Condens. Matter 2016, 28 (50), 503003. García-Abuín, A.; Gomez-Díaz, D.; Lopez, A. B.; Navaza, J. M.; Rumbo, A., NMR characterization of carbon dioxide chemical absorption with monoethanolamine, diethanolamine, and triethanolamine. Ind. Eng. Chem. Res. 2013, 52 (37), 13432-13438. Becke, A. D., Perspective: Fifty years of density-functional theory in chemical physics. J, Chem. Phys. 2014, 140 (18), 18A301. Richner, G.; Puxty, G., Assessing the Chemical Speciation during CO2 Absorption by Aqueous Amines Using in Situ FTIR. Ind. Eng. Chem. Res. 2012, 51 (44), 14317-14324. Jackson, P.; Robinson, K.; Puxty, G.; Attalla, M., In situ Fourier Transform-Infrared (FTIR) analysis of carbon dioxide absorption and desorption in amine solutions. Energy Procedia 2009, 1 (1), 985-994. Robinson, K.; McCluskey, A.; Attalla Moetaz, I., An FTIR Spectroscopic Study on the Effect of Molecular Structural Variations on the CO2 Absorption Characteristics of Heterocyclic Amines. ChemPhysChem 2011, 12 (6), 1088-1099. Bhugun, I.; Lexa, D.; Savéant, J.-M., Catalysis of the Electrochemical Reduction of Carbon Dioxide by Iron(0) Porphyrins. Synergistic Effect of Lewis Acid Cations. The Journal of Physical Chemistry 1996, 100 (51), 19981-19985. Westphal, E.; Pliego, J. R., Absolute solvation free energy of Li+ and Na+ ions in dimethyl sulfoxide solution: A theoretical ab initio and cluster-continuum model study. J. Chem. Phys. 2005, 123 (7), 074508. Guchik, I. V.; Frolov, Y. L.; Shagun, V. A.; Vashchenko, A. V.; Trofimov, B. A., Quantum-Chemical Analysis of Coordination of Dimethylsulfoxide Molecules to Sodium and Potassium Cations. J. Struct. Chem. 2004, 45 (1), 41-46. Siddique, A. A.; Dixit, M. K.; Tembe, B. L., Solvation structure and dynamics of potassium chloride ion pair in dimethyl sulfoxide–water mixtures. Journal of Molecular Liquids 2013, 188, 5-12. Malhotra, D.; Koech, P. K.; Heldebrant, D. J.; Cantu, D. C.; Zheng, F.; Glezakou, V.-A.; Rousseau, R., Reinventing Design Principles for Developing Low-Viscosity Carbon


24

Page 25 of 26 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60


47.

48. 49.

Dioxide-Binding Organic Liquids for Flue Gas Clean Up. ChemSusChem 2017, 10 (3), 636-642. Cantu, D. C.; Lee, J.; Lee, M.-S.; Heldebrant, D. J.; Koech, P. K.; Freeman, C. J.; Rousseau, R.; Glezakou, V.-A., Dynamic Acid/Base Equilibrium in Single Component Switchable Ionic Liquids and Consequences on Viscosity. J. Phys. Chem. Lett. 2016, 7 (9), 1646-1652. Luo, X. Y.; Fan, X.; Shi, G. L.; Li, H. R.; Wang, C. M., Decreasing the viscosity in CO 2 capture by amino-functionalized ionic liquids through the formation of intramolecular hydrogen bond. J. Phys. Chem. B 2016, 120 (10), 2807-2813. Towns, J.; Cockerill, T.; Dahan, M.; Foster, I.; Gaither, K.; Grimshaw, A.; Hazlewood, V.; Lathrop, S.; Lifka, D.; Peterson, G. D.; Roskies, R.; Scott, J. R.; Wilkins-Diehr, N., XSEDE: Accelerating Scientific Discovery. Computing in Science & Engineering 2014, 16 (5), 62-74.


25


A

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 26 of 26

TOC Graphic: O

B

Na+

+

OH

K

Li + Lithium Carbamate

CO2

B

O

Lean Amine, A2CO3

neCarbon electrode Carbamic Acid

OA Ammonium Perchlorate Lithium Carbamate -

Carbamic Acid

Li+ Na+ K+

+

Ammonium Perchlorate


26

Promoting Amine-Activated Electrochemical CO2 Conversion with

Recommend Documents